About

Florian Zettelmeyer is the Nancy L. Ertle Professor of Marketing at the Kellogg School of Management at Northwestern University. He specializes in evaluating the effects of analytics and artificial intelligence on firms, with research interests including marketing analytics, digital advertising, consumer search and uncertainty, industrial organization, pricing, and environmental economics. He is the founder and director of the Program on Data Analytics at Kellogg, where he leads efforts to advance understanding and application of data-driven insights in business. Professor Zettelmeyer received his PhD from the Massachusetts Institute of Technology and has held academic positions at the Haas School of Business at the University of California at Berkeley and the Simon Graduate School of Business Administration at the University of Rochester. He has been recognized with numerous teaching awards and was voted 'Outstanding Professor of the Year' by Kellogg MBA students. In addition to his academic role, he serves as the Chief Economist for Ads and Entertainment at Amazon, leading a team of over 100 economists, data scientists, and engineers to develop innovative ideas for Amazon's advertising and entertainment businesses.

Research topics

• Computer Science
• Business
• Marketing
• Advertising
• Data science
• World Wide Web
• Geography
• Finance
• Industrial organization

Selected publications

• Predicted Incrementality by Experimentation (PIE) for Ad Measurement

  SSRN Electronic Journal · 2026-01-01 · 2 citations

  preprintOpen accessSenior author
  PublisherDOI

• Amazon Ads Multi-Touch Attribution

  ArXiv.org · 2025-08-11

  preprintOpen access

  Amazon's new Multi-Touch Attribution (MTA) solution allows advertisers to measure how each touchpoint across the marketing funnel contributes to a conversion. This gives advertisers a more comprehensive view of their Amazon Ads performance across objectives when multiple ads influence shopping decisions. Amazon MTA uses a combination of randomized controlled trials (RCTs) and machine learning (ML) models to allocate credit for Amazon conversions across Amazon Ads touchpoints in proportion to their value, i.e., their likely contribution to shopping decisions. ML models trained purely on observational data are easy to scale and can yield precise predictions, but the models might produce biased estimates of ad effects. RCTs yield unbiased ad effects but can be noisy. Our MTA methodology combines experiments, ML models, and Amazon's shopping signals in a thoughtful manner to inform attribution credit allocation.

  PublisherOA PDFDOI

• Close Enough? A Large-Scale Exploration of Non-Experimental Approaches to Advertising Measurement

  arXiv (Cornell University) · 2022-01-18 · 25 citations

  preprintOpen accessSenior author

  Despite their popularity, randomized controlled trials (RCTs) are not always available for the purposes of advertising measurement. Non-experimental data is thus required. However, Facebook and other ad platforms use complex and evolving processes to select ads for users. Therefore, successful non-experimental approaches need to "undo" this selection. We analyze 663 large-scale experiments at Facebook to investigate whether this is possible with the data typically logged at large ad platforms. With access to over 5,000 user-level features, these data are richer than what most advertisers or their measurement partners can access. We investigate how accurately two non-experimental methods -- double/debiased machine learning (DML) and stratified propensity score matching (SPSM) -- can recover the experimental effects. Although DML performs better than SPSM, neither method performs well, even using flexible deep learning models to implement the propensity and outcome models. The median RCT lifts are 29%, 18%, and 5% for the upper, middle, and lower funnel outcomes, respectively. Using DML (SPSM), the median lift by funnel is 83% (173%), 58% (176%), and 24% (64%), respectively, indicating significant relative measurement errors. We further characterize the circumstances under which each method performs comparatively better. Overall, despite having access to large-scale experiments and rich user-level data, we are unable to reliably estimate an ad campaign's causal effect.

  PublisherOA PDFDOI

• Close Enough? A Large-Scale Exploration of Non-Experimental Approaches to Advertising Measurement

  Marketing Science · 2022 · 80 citations

  Senior authorCorresponding
  • Computer Science
  • Advertising
  • Computer Science

  A large-scale comparison of experimental advertising effects and those obtained using two state-of-the-art methods.

  PublisherDOI

• How Market Power Affects Dynamic Pricing: Evidence from Inventory Fluctuations at Car Dealerships

  Management Science · 2021-10-14 · 15 citations

  articleSenior author

  This paper investigates empirically the effect of market power on dynamic pricing in the presence of inventories. Our setting is the auto retail industry; we analyze how automotive dealerships adjust prices to inventory levels under varying degrees of market power. We first establish that inventory fluctuations create scarcity rents for cars that are in short supply. We then show that dealers’ ability to adjust prices in response to inventory depends on their market power, that is, the quantity of substitute inventory in their selling area. Specifically, we show that the slope of the price–inventory relationship (higher inventory lowers prices) is significantly steeper when dealers find themselves in a situation of high rather than low market power. A dealership with high market power moving from a situation of inventory shortage to a median inventory level lowers transaction prices by about 0.57% ceteris paribus, corresponding to 32.5% of dealers’ average per-vehicle profit margin or $145.6 on the average car. Conversely, when competition is more intense, moving from inventory shortage to a median inventory level lowers transaction prices by about 0.35% ceteris paribus, corresponding to 20.2% of dealers’ average per-vehicle profit margin or $90.9. To our knowledge, we are the first to empirically show that market power affects firms’ ability to dynamically price. This paper was accepted by Juanjuan Zhang, marketing.

  PublisherDOI

• Open Negotiation: The Back-End Benefits of Salespeople’s Transparency in the Front End

  Journal of Marketing Research · 2020 · 36 citations

  Senior authorCorresponding
  • Computer Science
  • Business
  • Marketing

  Negotiations today are less likely to be characterized by information asymmetry—the notion that buyers are less informed than sellers—due to the amount of information available to buyers. A number of industries have reacted to this change by shifting their attention to earning profits in aftermarkets: products and services that augment the main purchase (e.g., add-ons, insurance, financing, service and maintenance). In these aftermarkets, firms often retain an information advantage, even if information asymmetries are eliminated from the main purchase. This has given rise to an interesting setting untapped by prior research: information “symmetry” in the front end (main purchase) and information “asymmetry” in the back end (aftermarket). The authors argue that symmetry in the front end provides an opportunity to build trust, as the knowledgeable customer can verify the information disclosed by the seller. In an observational study in the automotive industry, the authors find that customers to whom the salesperson revealed the cost of a car at the beginning of the negotiation spent significantly more in the back end than others. As corroborated in subsequent studies, this effect holds only when cost is disclosed at the beginning of the negotiation and when customers can verify the cost information.

  PublisherDOI

• A Comparison of Approaches to Advertising Measurement: Evidence from Big Field Experiments at Facebook

  Marketing Science · 2019-03-01 · 268 citations

  article

  Observational methods often fail to accurately recover the treatment effects generated from randomized advertising experiments on Facebook.

  PublisherDOI

• A Comparison of Approaches to Advertising Measurement: Evidence from Big Field Experiments at Facebook

  SSRN Electronic Journal · 2017-01-01 · 71 citations

  articleOpen access
  PublisherDOI

• CDK Digital Marketing: Addressing Channel Conflict with Data Analytics

  Kellogg School of Management Cases · 2017-01-20

  article1st authorCorresponding

  Four years into a five-year contract with General Motors to be the exclusive website vendor to its U.S. network of more than 4,000 dealers, CDK Digital faced a crucial contract renewal at the end of 2012. The case follows Melissa McCann, director of strategic marketing, and Chris Reed, CMO, as they prepared for a critical meeting in July 2011: a presentation to the customer relationship management (CRM) subcommittee of the Chevrolet dealer council. Although GM dealers, like all auto dealers in the United States, were independent franchisees, GM saw the renewal of CDK Digital's exclusive contract as a collaborative decision between dealers and GM. According to Ed Vogt, GM's executive in charge of the renewal, if the dealer councils said no, the contract would not be renewed. This case challenges students to use CDK's big data and analytics capabilities to address the inherent conflict between dealers and manufacturers: when marketing to potential customers, manufacturers wanted consistency across dealer websites to maximize sales of their targeted brands, while dealers wanted flexibility to sell what they had in inventory. After analyzing the case, students will be able to:

  PublisherDOI

• Who is exposed to gas prices? How gasoline prices affect automobile manufacturers and dealerships

  DSpace@MIT (Massachusetts Institute of Technology) · 2016-05-01 · 1 citations

  articleOpen access
  Publisher

Recent grants

• Collaborative Research: Incentive promotions in the US automotive industry

  NSF · $152k · 2006–2012

• Collaborative Research: Incentive promotions in the US automotive industry

  NSF · $137k · 2012–2013

Frequent coauthors

• Meghan R. Busse

  44 shared

• Jorge Silva‐Risso

  University of California, Riverside

  38 shared

• Fiona Scott Morton

  Heinrich Heine University Düsseldorf

  24 shared

• Christopher R. Knittel

  National Bureau of Economic Research

  17 shared

• Duncan Simester

  16 shared

• Fiona M. Scott Morton

  Yale University

  16 shared

• Steven Tadelis

  University of California, Berkeley

  8 shared

• Brett R. Gordon

  7 shared

Labs

• Program on Data Analytics at KelloggPI

Awards & honors

• John D.C. Little Award for best marketing paper published in…
• John D.C. Little Award for best marketing paper published in…
• 2021 Sales SIG Excellence in Research Award, American Market…
• Finalist for L. G. Lavengood Outstanding Professor of the Ye…
• Finalist for L. G. Lavengood Outstanding Professor of the Ye…